In renewing tissues such as stratified squamous epithelia (e.g., epidermis, oral mucosa) cell replication occurs continually, with the concomitant potential to acquire new mutations. Thus the probability increases with time and age that a constituent cell (keratinocyte) will acquire a combination of heritable alterations that permit it to evade normal restraints and become a cancer cell (squamous cell carcinoma (SCC)). Limiting the probability that a cell might progressively accumulate oncogenic mutations is an internal mechanism, activated by progressive telomere shortening, that imposes a state of replicative senescence on a cell after it has undergone a certain number of divisions. Earlier studies identified p53 as an essential enforcer of this telomere length-sensitive growth arrest mechanism. Recently we and others have found that another cell cycle inhibitor, INK4a (p16), can enforce a senescence arrest on keratinocytes and other cell types. p16 expression is activated by a sensor that apparently detects some aspect of cell age or physiological status unrelated to telomere length. Keratinocytes engineered to express telomerase, such that p53-dependent arrest is not triggered, cannot divide rapidly as immortalized cells unless they undergo mutation to evade the p16-enforced arrest system. Potentially intermediate stages of neoplastic progression are identifiable in epithelia as lesions of abnormal appearing cells (dysplasias), which are often found to have acquired some p16 and/or p53 mutations. At present there are no reliable ways to predict which lesions are premalignant (i.e., which will progress rapidly to SCC) and which will remain indolent or regress. We hypothesize that keratinocytes become premalignant when they have acquired mutations permitting them to bypass both p16- and p53-dependent senescence mechanisms and become extended lifespan, yet still non-immortal, cells. We have cultured cells having these characteristics from a dysplastic oral lesion and we have constructed an experimental model cell line from normal human keratinocytes by stably expressing proteins that block the activity of endogenous p16 and p53. We propose to test these cell lines, and others we generate using a combination of gene transduction and spontaneous mutant selection strategies, for susceptibility to induction of telomerase expression by transcriptional transactivators such as myc, for their ability to generate spontaneous immortalized variants after dividing to crisis, and for their tumorigenic behavior in athymic mice. We will also begin to identify and characterize other types of mutations that can result in inactivation of the p16- dependent mechanism in keratinocytes.